“We could have shipped it.

The discussion of Episode 3 begins at 1h 52m, with concept art and eventually in-game footage.

The in-game footage is all taken from test arenas and tech demos, not finished parts of the game.

A liquid blob moves around a tree in a Half-Life 2: Epsiode 3 tech demo.

There are two main features of the episode discussed.

They are blobs, like liquid droplets, which can change shape, separate, and sluice around environments.

They could absorb other enemies or physics objects, and squeeze (or fall) through grates.

Cover image for YouTube video

They certainly look cool, though.

Y’know, it’s like - we have new features?

What kind of story can we do with these now?”

It just seemed like a fun thing to do… until I did it."

Several Valve employees in the docuemntary offer their own takes on why they never continued with Episode 3.

“you’re able to’t get lazy and say, ‘We’re moving the story forward.’

That’s copping out of your obligation to gamers.

Yes, of course they love the story, they love many, many aspects of it.

It wouldn’t have been that hard,” says Gabe Newell.

“The failure was, my personal failure was being stumped.

I couldn’t figure out why doing Episode 3 was pushing anything forward.”

I don’t think it’s lazy to resolve a story you ended on a cliffhanger.

The whole anniversary documentary is worth watching, if you’re a Half-Life obsessive like me.

There’s brief footage in there of Arkane’s abandoned Ravenholm project, near the end, for example.